skills/user-story-delivery/SKILL.md
Orchestrate the complete delivery workflow for one GitHub user story: implement it with the user-story-implementer skill, review the resulting PR with the user-story-reviewer skill, address reviewer feedback, and repeat until the PR is approved, merged, or blocked. You MUST use this skill when asked to "implement and review a user story", "run the full user story workflow", "deliver USERST-001", "complete USERST-001 end to end", or "run implementation and review together".
npx skillsauth add eho/agent-skills user-story-deliveryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
You are acting as the coordinator for one complete user story delivery cycle. Your job is to compose two specialist workflows:
user-story-implementer: implements exactly one story, verifies it, commits, pushes, and creates or updates a PR.user-story-reviewer: reviews the PR against the original issue and either requests changes, fixes a small issue, approves, comments, or merges according to its own workflow.Keep implementation and review responsibilities separate. Do not duplicate either specialist skill's detailed workflow here. The value of this skill is orchestration: clear handoffs, bounded review-fix loops, and a final delivery status the user can trust.
PREREQUISITE: The GitHub CLI (gh) MUST be installed and fully authenticated (gh auth login) because the specialist skills rely on it.
This workflow is designed to use subagents so implementation and review remain independent. If the current runtime requires explicit user permission before starting subagents, and the user's request did not clearly ask for subagents, delegation, or the full implement-review workflow, ask for confirmation before starting. A request such as "use the user-story-delivery skill to run the full workflow" is enough to proceed because this skill's purpose is delegation.
Identify the Target Story
Start Implementation
user-story-implementer skill.## Implementation Handoff
- Story ID:
- Issue:
- Branch:
- PR:
- Verification:
- Known residual risk:
- Blocked: yes/no
Blocked: yes, stop and relay the blocker. Do not start review.Start Review
user-story-reviewer skill.## Review Handoff
- Story ID:
- PR:
- Decision: Request changes | Fix small issue | Approve | Comment only | Merge
- Blocking findings:
- Reviewer-fixed commits:
- Required follow-up:
- Verification:
Handle Review Decision
Decision is Approve, Comment only, or Merge, stop the loop and provide the final delivery report.Decision is Fix small issue, check that the reviewer pushed the fix and then continue to a follow-up review.Decision is Request changes, address the feedback on the existing PR branch.Address Requested Changes
user-story-implementer skill to revise the existing PR branch.Implementation Handoff format, plus a short list of review findings addressed.Repeat Review
Use concise prompts that include the handoff requirements. Adapt the target story details as needed.
Use the user-story-implementer skill to implement exactly one user story: <story-id-or-selection-rule>.
You are not alone in the codebase. Do not revert unrelated changes. Follow the implementer workflow, verify the acceptance criteria, commit, push, and create or update the PR.
Return this final handoff:
## Implementation Handoff
- Story ID:
- Issue:
- Branch:
- PR:
- Verification:
- Known residual risk:
- Blocked: yes/no
Use the user-story-reviewer skill to review this implemented story.
Story: <story-id>
PR: <pr-number-or-url>
Complete the reviewer workflow. If there are no blocking findings, approve, comment, or merge according to the reviewer skill's rules. If there are findings, request changes unless the reviewer skill permits fixing a small issue directly.
Return this final handoff:
## Review Handoff
- Story ID:
- PR:
- Decision: Request changes | Fix small issue | Approve | Comment only | Merge
- Blocking findings:
- Reviewer-fixed commits:
- Required follow-up:
- Verification:
Use the user-story-implementer skill to revise the existing PR for this story.
Story: <story-id>
PR: <pr-number-or-url>
Reviewer feedback to address:
<findings>
You are not alone in the codebase. Do not revert unrelated changes. Check out and update the existing PR branch, address the review findings, verify the affected behavior, commit, and push to the same PR.
Return this final handoff:
## Implementation Handoff
- Story ID:
- Issue:
- Branch:
- PR:
- Review findings addressed:
- Verification:
- Known residual risk:
- Blocked: yes/no
When the workflow stops, report:
## Delivery Status
- Story:
- Issue:
- PR:
- Final decision:
- Review cycles:
- Implementation summary:
- Verification:
- Remaining blockers or residual risk:
If the story was merged, say so explicitly. If it was approved but not merged, explain why. If it stopped because of blockers or remaining findings, lead with the unresolved items and include the exact next action needed.
documentation
Compact the current conversation into a handoff document for another agent to pick up.
tools
--- name: expo-ios-agent-device description: Drive the Kotoba Expo iOS simulator effectively with agent-device. Use when verifying iOS UI behavior, testing Expo dev-client flows, seeding simulator app state, diagnosing accessibility selectors, or automating simulator QA for this repo. --- # Expo iOS Agent Device Use this skill when automating Kotoba on the iOS simulator with `agent-device`. ## Preflight Run these before planning commands: ```sh agent-device --version agent-device help workf
development
Create or update a DESIGN.md design system specification for websites, apps, prototypes, or product designs. Use when the user asks to generate a DESIGN.md, design system spec, UI specification, frontend design source of truth, Stitch/Google Design.md-style document, or agent-ready design tokens and component guidelines from images, descriptions, screenshots, mockups, brand notes, or raw product requirements.
documentation
Run the review-revision loop for an existing design doc: call design-doc-reviewer, address Critical Gaps and Minor Issues, repeat until clean, then mark the doc Revised for feature-delivery. Use when asked to review and revise, repeat until no gaps remain, prepare a design doc for feature delivery, or start a reviewer subagent and address feedback. Use design-doc-reviewer for one-time read-only critique.